Search Results (1,534)

Ensuring global food security requires accurate and robust solutions for crop health monitoring, weed detection, and large-scale land-cover classification. To this end, we propose AgroVisionNet, a hybrid deep learning framework that integrates Convolutional Neural Networks (CNNs) for local feature extraction, Vision Transformers (ViTs) for capturing long-range global dependencies, and Graph Neural Networks (GNNs) for modeling spatial relationships between image regions. The framework was evaluated on five diverse benchmark datasets—PlantVillage (leaf-level disease detection), Agriculture-Vision (field-scale anomaly segmentation), BigEarthNet (satellite-based land-cover classification), UAV Crop and Weed (weed segmentation), and EuroSAT (multi-class land-cover recognition). Across these datasets, AgroVisionNet consistently outperformed strong baselines including ResNet-50, EfficientNet-B0, ViT, and Mask R-CNN. For example, it achieved 97.8% accuracy and 95.6% IoU on PlantVillage, 94.5% accuracy on Agriculture-Vision, 92.3% accuracy on BigEarthNet, 91.5% accuracy on UAV Crop and Weed, and 96.4% accuracy on EuroSAT. These results demonstrate the framework’s robustness across tasks ranging from fine-grained disease detection to large-scale anomaly mapping. The proposed hybrid approach addresses persistent challenges in agricultural imaging, including class imbalance, image quality variability, and the need for multi-scale feature integration. By combining complementary architectural strengths, AgroVisionNet establishes a new benchmark for deep learning applications in precision agriculture. Full article

Reliable spatio-temporal rainfall prediction is a key element in disaster mitigation and water resource management in dynamic tropical regions such as the Bengawan Solo River Watershed. However, high climate variability and data limitations often pose significant challenges to the accuracy of conventional prediction models. This study introduces an innovative approach by applying ensemble stacking, which combines machine learning models such as Random Forest (RF), Extreme Gradient Boosting (XGB), Support Vector Regression (SVR), Multi-Layer Perceptron (MLP), Light Gradient-Boosting Machine (LGBM) and deep learning models like Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Temporal Convolutional Networks (TCN), Convolutional Neural Network (CNN), and Transformer architecture based on monthly Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) data (1981–2024). The novelty of this research lies in the systematic exploration of various model combination scenarios—both classical and deep learning and the evaluation of their performance in projecting rainfall for 2025–2030. All base models were trained on the 1981–2019 period and validated with data from the 2020–2024 period, while ensemble stacking was developed using a linear regression meta-learner. The results show that the optimal ensemble scenario reduces the MAE to 53.735 mm, the RMSE to 69.242 mm, and increases the R² to 0.795826—better than all individual models. Spatial and temporal analyses also indicate consistent model performance at most locations and times. Annual rainfall projections for 2025–2030 were then interpolated using IDW to generate a spatio-temporal rainfall distribution map. The improved accuracy provides a strong scientific basis for disaster preparedness, flood and drought management, and sustainable water planning in the Bengawan Solo River Watershed. Beyond this case, the approach demonstrates significant transferability to other climate-sensitive and data-scarce regions. Full article

Forest fires pose significant threats to ecosystems, human life, and the global climate, necessitating rapid and reliable detection systems. Traditional fire detection approaches, including sensor networks, satellite monitoring, and centralized image analysis, often suffer from delayed response, high false positives, and limited deployment in remote areas. Recent deep learning-based methods offer high classification accuracy but are typically computationally intensive and unsuitable for low-power, real-time edge devices. This study presents an autonomous, edge-based forest fire and smoke detection system using a lightweight MobileNetV2 convolutional neural network. The model is trained on a balanced dataset of fire, smoke, and non-fire images and optimized for deployment on resource-constrained edge devices. The system performs near real-time inference, achieving a test accuracy of 97.98% with an average end-to-end prediction latency of 0.77 s per frame (approximately 1.3 FPS) on the Raspberry Pi 5 edge device. Predictions include the class label, confidence score, and timestamp, all generated locally without reliance on cloud connectivity, thereby enhancing security and robustness against potential cyber threats. Experimental results demonstrate that the proposed solution maintains high predictive performance comparable to state-of-the-art methods while providing efficient, offline operation suitable for real-world environmental monitoring and early wildfire mitigation. This approach enables cost-effective, scalable deployment in remote forest regions, combining accuracy, speed, and autonomous edge processing for timely fire and smoke detection. Full article

To improve the physical consistency and interpretability of traditional drought indices, this study proposes a drought assessment model that couples physically based constraints with data-driven approaches, leading to the development of a Multivariate Drought Index (MDI). The model employs convolutional neural networks to achieve physically consistent downscaling, thereby obtaining a high-resolution Normalized Difference Water Index (NDWI), Temperature Vegetation Dryness Index (TVDI), Vegetation Condition Index (VCI), and Temperature Condition Index (TCI). Objective weights are determined using the Criteria Importance Through Intercriteria Correlation method, while random forest and Shapley Additive Explanations are integrated for nonlinear interpretation and physics-guided calibration, forming an ensemble framework that incorporates multi-source and multi-scale factors. Validation with multi-source data from 2000 to 2024 in the major maize-growing areas of Heilongjiang Province demonstrates that MDI outperforms single indices and the Vegetation Health Index (VHI), achieving a correlation coefficient (r = 0.87), coefficient of determination (R² = 0.87), RMSE (0.08), and classification accuracy (87.4%). During representative drought events, MDI identifies signals 16–20 days earlier than the Standardized Precipitation Evapotranspiration Index (SPEI) and the Soil Moisture Condition Index (SMCI), and effectively captures localized drought patches at a 250 m scale. Feature importance analysis indicates that the NDWI and TVDI are consistently identified as dominant factors across all three methods, aligning physically interpretable analysis with statistical contribution. Long-term risk zoning reveals that the central–western region of the study area constitutes a high-risk zone, accounting for 42.6% of the total. This study overcomes the limitations of single indices by integrating physical consistency with the advantages of data-driven methods, achieving refined spatiotemporal characterization and enhanced overall performance, while also demonstrating potential for application across different crops and regions. Full article

Classifying brain tumour transcriptomic data is crucial for precision medicine but remains challenging due to high dimensionality and limited interpretability of conventional models. This study benchmarks three image-based deep learning approaches, DeepInsight, Fotomics, and a novel saliency-guided convolutional neural network (CNN), for transcriptomic classification. DeepInsight utilises dimensionality reduction to spatially arrange gene features, while Fotomics applies Fourier transforms to encode expression patterns into structured images. The proposed method transforms each single-cell gene expression profile into an RGB image using PCA, UMAP, or t-SNE, enabling CNNs such as ResNet to learn spatially organised molecular features. Gradient-based saliency maps are employed to highlight gene regions most influential in model predictions. Evaluation is conducted on two biologically and technologically different datasets: single-cell RNA-seq from glioblastoma GSM3828672 and bulk microarray data from medulloblastoma GSE85217. Outcomes demonstrate that image-based deep learning methods, particularly those incorporating saliency guidance, provide a robust and interpretable framework for uncovering biologically meaningful patterns in complex high-dimensional omics data. For instance, ResNet-18 achieved the highest accuracy of 97.25% on the GSE85217 dataset and 91.02% on GSM3828672, respectively, outperforming other baseline models across multiple metrics. Full article

Recently, deep learning models such as convolutional neural networks (CNNs), convolutional autoencoders (CAEs), CNN-based support vector machines (SVMs), YOLO, fully convolutional networks (FCNs), fully convolutional data descriptions (FCDDs) and so on have been applied to defect detections and anomaly detections of various kinds of industrial products, materials and systems. In those models, downsampled images, including target features, are used for training and testing. On the other hand, although various types of anomaly detection systems based on time series data such as sounds and vibrations are also applied to manufacturing processes, complicated conversions to the frequency domain are basically needed in conventional approaches. This paper addresses an important industrial problem for detecting anomalies in machine tools at low cost using audio data. Intelligent anomaly diagnosis systems for computer numerical control (CNC) machine tools are considered and proposed, in which raw time-series data without the need of conversion to the frequency domain can be directly used for training and testing. As for the NN models for comparison, conventional shallow NN, RNN and 1D CNN are designed and trained using the nine kinds of mechanical sounds. Classification results of test sound block (SB) data by the three models are shown. Then, an autoencoder (AE) is designed and considered for the identifier by training it using only normal SB data of a machine tool. One of the technical needs in dealing with time-series data such as SB data by NNs is how to clearly visualize and understand anomalous regions in concurrence with identification. Finally, we propose the SB data-based FCDD model to meet this need. Basic performance of the SB data-based FCDD model is evaluated in terms of anomaly detection and concurrent visualization of understanding. Full article

This study develops an innovative unmanned aerial vehicle (UAV)-based intelligent system for winter wheat yield prediction, addressing the inefficiencies of traditional manual counting methods (with approximately 15% error rate) and enabling quantitative analysis of water–fertilizer interactions. By integrating an enhanced Faster Region-Based Convolutional Neural Network (Faster R-CNN) architecture with multi-source data fusion and machine learning, the system significantly improves both spike detection accuracy and yield forecasting performance. Field experiments during the 2022–2023 growing season captured high-resolution multispectral imagery for varied irrigation regimes and fertilization treatments. The optimized detection model incorporates ResNet-50 as the backbone feature extraction network, with residual connections and channel attention mechanisms, achieving a mean average precision (mAP) of 91.2% (calculated at IoU threshold 0.5) and 88.72% recall while reducing computational complexity. The model outperformed YOLOv8 by a statistically significant 2.1% margin (p < 0.05). Using model-generated spike counts as input, the random forest (RF) model regressor demonstrated superior yield prediction performance (R² = 0.82, RMSE = 324.42 kg·ha⁻¹), exceeding the Partial Least Squares Regression (PLSR) (R² +46%, RMSE-44.3%), Least Squares Support Vector Machine (LSSVM) (R² + 32.3%, RMSE-32.4%), Support Vector Regression (SVR) (R² + 30.2%, RMSE-29.6%), and Backpropagation (BP) Neural Network (R²+22.4%, RMSE-24.4%) models. Analysis of different water–fertilizer treatments revealed that while organic fertilizer under full irrigation (750 m³ ha⁻¹) conditions achieved maximum yield benefit (13,679.26 CNY·ha⁻¹), it showed relatively low water productivity (WP = 7.43 kg·m⁻³). Conversely, under deficit irrigation (450 m³ ha⁻¹) conditions, the 3:7 organic/inorganic fertilizer treatment achieved optimal WP (11.65 kg m⁻³) and WUE (20.16 kg∙ha⁻¹∙mm⁻¹) while increasing yield benefit by 25.46% compared to organic fertilizer alone. This research establishes an integrated technical framework for high-throughput spike monitoring and yield estimation, providing actionable insights for synergistic water–fertilizer management strategies in sustainable precision agriculture. Full article

Raman spectroscopy is a rapid, label-free, and non-destructive technique for probing molecular structures, making it a powerful tool for clinical pathogen identification. However, interpreting its complex spectral data remains challenging. In this study, we evaluate and compare a suite of machine learning models—including Support Vector Machines (SVM), XGBoost, LightGBM, Random Forests, k-nearest Neighbors (k-NN), Convolutional Neural Networks (CNNs), and fully connected Neural Networks—with and without Principal Component Analysis (PCA) for dimensionality reduction. Using Raman spectral data from 30 clinically important bacterial and fungal species that collectively account for over 90% of human infections in hospital settings, we conducted rigorous hyperparameter tuning and assessed model performance based on accuracy, precision, recall, and F1-score. The SVM with an RBF kernel combined with PCA emerged as the top-performing model, achieving the highest accuracy (0.9454) and F1-score (0.9454). Ensemble methods such as LightGBM and XGBoost also demonstrated strong performance, while CNNs provided competitive results among deep learning approaches. Importantly, interpretability was achieved via SHAP (Shapley Additive exPlanations), which identified class-specific Raman wavenumber regions critical to prediction. These interpretable insights, combined with strong classification performance, underscore the potential of explainable AI-driven Raman analysis to accelerate clinical microbiology diagnostics, optimize antimicrobial therapy, and improve patient outcomes. Full article

Background/Objectives: The clinical management of adolescent idiopathic scoliosis (AIS) is hindered by the inability to accurately predict curve progression. Although skeletal maturity and the initial Cobb angle are established predictors of progression, their combined predictive accuracy remains limited. This study aimed to develop a robust and interpretable artificial intelligence (AI) system using deep learning (DL) models to predict the progression of scoliosis using only standing frontal radiographs. Methods: We conducted a multicenter study involving 542 patients with AIS. After excluding 52 borderline progression cases (6–9° progression in the Cobb angle), 294 and 196 patients were assigned to progression (≥10° increase) and non-progression (≤5° increase) groups, respectively, considering a 2-year follow-up. Frontal whole spinal radiographs were preprocessed using histogram equalization and divided into two regions of interest (ROIs) (ROI 1, skull base–femoral head; ROI 2, C7–iliac crest). Six pretrained DL models, including convolutional neural networks (CNNs) and transformer-based models, were trained on the radiograph images. Gradient-weighted class activation mapping (Grad-CAM) was further performed for model interpretation. Results: Ensemble models outperformed individual ones, with the average ensemble model achieving area under the curve (AUC) values of 0.769 for ROI 1 and 0.755 for ROI 2. Grad-CAM revealed that the CNNs tended to focus on the local curve apex, whereas the transformer-based models demonstrated global attention across the spine, ribs, and pelvis. Models trained on ROI 2 performed comparably with respect to those using ROI 1, supporting the feasibility of image standardization without a loss of accuracy. Conclusions: This study establishes the clinical potential of transformer-based DL models for predicting the progression of scoliosis using only plain radiographs. Our multicenter approach, high AUC values, and interpretable architectures support the integration of AI into clinical decision-making for the early treatment of AIS. Full article

Under the context of “Energy Conservation and Emission Reduction,” low-sulfur fuel has become widely adopted in maritime operations, posing significant challenges to cylinder lubrication systems. Traditional oil injection strategies, heavily reliant on manual experience, suffer from instability and high costs. To address this, a lightweight image retrieval model for cylinder lubrication is proposed, leveraging deep learning and computer vision to support oiling decisions based on visual features. The model comprises three components: a backbone network, a feature enhancement module, and a similarity retrieval module. Specifically, EfficientNetB0 serves as the backbone for efficient feature extraction under low computational overhead. MobileViT Blocks are integrated to combine local feature perception of Convolutional Neural Networks (CNNs) with the global modeling capacity of Transformers. To further improve receptive field and multi-scale representation, Receptive Field Blocks (RFB) are introduced between the components. Additionally, the Convolutional Block Attention Module (CBAM) attention mechanism enhances focus on salient regions, improving feature discrimination. A high-quality image dataset was constructed using WINNING’s large bulk carriers under various sea conditions. The experimental results demonstrate that the EfficientNetB0 + RFB + MobileViT + CBAM model achieves excellent performance with minimal computational cost: 99.71% Precision, 99.69% Recall, and 99.70% F1-score—improvements of 11.81%, 15.36%, and 13.62%, respectively, over the baseline EfficientNetB0. With only a 0.3 GFLOP and 8.3 MB increase in model size, the approach balances accuracy and inference efficiency. The model also demonstrates good robustness and application stability in real-world ship testing, with potential for further adoption in the field of intelligent ship maintenance. Full article

Timely identification and accurate delineation of ultra-early ischemic stroke lesions in non-contrast computed tomography (CT) scans of the human brain are of paramount importance for prompt medical intervention and improved patient outcomes. In this study, we propose a deep learning-driven methodology specifically designed for segmenting ultra-early ischemic regions, with a particular emphasis on both the ischemic core and the surrounding penumbra during the initial stages of stroke progression. We introduce a lightweight preprocessing model based on convolutional filtering techniques, which enhances image clarity while preserving the structural integrity of medical scans, a critical factor when detecting subtle signs of ultra-early ischemic strokes. Unlike conventional preprocessing methods that directly modify the image and may introduce artifacts or distortions, our approach ensures the absence of neural network-induced artifacts, which is especially crucial for accurate diagnosis and segmentation of ultra-early ischemic lesions. The model employs predefined differentiable filters with trainable parameters, allowing for artifact-free and precision-enhanced image refinement tailored to the challenges of ultra-early stroke detection. In addition, we incorporated into the combined preprocessing pipeline a newly proposed trainable linear combination of pretrained image filters, a concept first introduced in this study. For model training and evaluation, we utilize a publicly available dataset of acute ischemic stroke cases, focusing on the subset relevant to ultra-early stroke manifestations, which contains annotated non-contrast CT brain scans from 112 patients. The proposed model demonstrates high segmentation accuracy for ultra-early ischemic regions, surpassing existing methodologies across key performance metrics. The results have been rigorously validated on test subsets from the dataset, confirming the effectiveness of our approach in supporting the early-stage diagnosis and treatment planning for ultra-early ischemic strokes. Full article

High-resolution remote sensing (HRRS) imagery enables the extraction of cropland information with high levels of detail, especially when combined with the impressive performance of deep convolutional neural networks (DCNNs) in understanding these images. Comprehending the factors influencing DCNNs’ performance in HRRS cropland extraction is of considerable importance for practical agricultural monitoring applications. This study investigates the impact of classifier selection and different training data characteristics on the HRRS cropland classification outcomes. Specifically, Gaofen-1 composite images with 2 m spatial resolution are employed for HRRS cropland extraction, and two county-wide regions with distinct agricultural landscapes in Shandong Province, China, are selected as the study areas. The performance of two deep learning (DL) algorithms (UNet and DeepLabv3+) and a traditional classification algorithm, Object-Based Image Analysis with Random Forest (OBIA-RF), is compared. Additionally, the effects of different band combinations, crop growth stages, and class mislabeling on the classification accuracy are evaluated. The results demonstrated that the UNet and DeepLabv3+ models outperformed OBIA-RF in both simple and complex agricultural landscapes, and were insensitive to the changes in band combinations, indicating their ability to learn abstract features and contextual semantic information for HRRS cropland extraction. Moreover, compared with the DL models, OBIA-RF was more sensitive to changes in the temporal characteristics. The performance of all three models was unaffected when the mislabeling error ratio remained below 5%. Beyond this threshold, the performance of all models decreased, with UNet and DeepLabv3+ showing similar performance decline trends and OBIA-RF suffering a more drastic reduction. Furthermore, the DL models exhibited relatively low sensitivity to the patch size of sample blocks and data augmentation. These findings can facilitate the design of operational implementations for practical applications. Full article

With the rapid proliferation of unmanned aerial vehicles (UAVs) and the associated rise in security concerns, there is a growing demand for robust detection and identification systems capable of operating reliably in complex electromagnetic environments. To address this challenge, this paper proposes a deep learning-based passive UAV detection and identification system leveraging radio frequency (RF) spectrograms. The system employs a high-resolution RF front-end comprising a multi-beam directional antenna and a wideband spectrum analyzer to scan the target airspace and capture UAV signals with enhanced spatial and spectral granularity. A YOLO-based detection module is then used to extract frequency hopping signal (FHS) regions from the spectrogram, which are subsequently classified by a convolutional neural network (CNN) to identify specific UAV models. Extensive measurements are carried out in both line-of-sight (LoS) and non-line-of-sight (NLoS) urban environments. The proposed system achieves over 96% accuracy in both detection and identification under LoS conditions. In NLoS conditions, it improves the identification accuracy by more than 15% compared with conventional full-spectrum CNN-based methods. These results validate the system’s robustness, real-time responsiveness, and strong practical applicability. Full article

The background of the article refers to the diagnosis of focal liver lesions (FLLs) through contrast-enhanced ultrasound (CEUS) based on the integration of spatial and temporal information. Traditional computer-aided diagnosis (CAD) systems predominantly rely on static images, which limits the characterization of lesion dynamics. This study aims to assess the effectiveness of Transformer-based architectures in enhancing CAD performance within the realm of liver pathology. The methodology involved a systematic comparison of deep learning models for the analysis of CEUS images and videos. For image-based classification, a Hybrid Transformer Neural Network (HTNN) was employed. It combines Vision Transformer (ViT) modules with lightweight convolutional features. For video-based tasks, we evaluated a custom spatio-temporal Convolutional Neural Network (CNN), a CNN with Long Short-Term Memory (LSTM), and a Video Vision Transformer (ViViT). The experimental results show that the HTNN achieved an outstanding accuracy of 97.77% in classifying various types of FLLs, although it required manual selection of the region of interest (ROI). The video-based models produced accuracies of 83%, 88%, and 88%, respectively, without the need for ROI selection. In conclusion, the findings indicate that Transformer-based models exhibit high accuracy in CEUS-based liver diagnosis. This study highlights the potential of attention mechanisms to identify subtle inter-class differences, thereby reducing the reliance on manual intervention. Full article

The concentration characteristics of SO₂, NO₂, O₃, and CO in the marine atmosphere are of great significance for understanding air–sea interactions and regional atmospheric chemical processes. However, due to the challenging conditions of marine monitoring, long-term continuous observational data remain scarce. To address this gap, this study proposes a Transfer Learning–Convolutional Neural Network (TL-CNN) model that integrates ERA5 meteorological data, EAC4 atmospheric composition reanalysis data, and ground-based observations through multi-source data fusion. During data preprocessing, the Data Interpolating Empirical Orthogonal Function (DINEOF), inverse distance weighting (IDW) spatial interpolation, and Gaussian filtering methods were employed to improve data continuity and consistency. Using ERA5 meteorological variables as inputs and EAC4 pollutant concentrations as training targets, a CNN-based inversion framework was constructed. Results show that the CNN model achieved an average coefficient of determination (R²) exceeding 0.80 on the pretraining test set, significantly outperforming random forest and deep neural networks, particularly in reproducing nearshore gradients and regional spatial distributions. After incorporating transfer learning and fine-tuning with station observations, the model inversion results reached an average R² of 0.72 against site measurements, effectively correcting systematic biases in the reanalysis data. Among the pollutants, the inversion of SO₂ performed relatively poorly, mainly because emission reduction trends from anthropogenic sources were not sufficiently represented in the reanalysis dataset. Overall, the TL-CNN model provides more accurate pollutant concentration fields for offshore regions with limited observations, offering strong support for marine atmospheric environment studies and assessments of marine ecological effects. It also demonstrates the potential of combining deep learning and transfer learning in atmospheric chemistry research. Full article